WIRELESS TELEGRAPHY. 289 



ble to signal with a g-iven amount of energy varies approximately with 

 the square of the height of the vertical wire, and with the square root 

 of the capacity of a plate, drum, or other form of capacity area which 

 may be placed at t '^ top of the wires. 



The law governiNij the relation of height and distance has alreadj- 

 been proved correct up to a distance of 85 miles. Many months ago 

 it was found possible to communicate from the North Haven. Poole, 

 to Alum Ba3% Isle of Wight, with a height of 75 feet, the distance being 

 18 miles. Later on tAvo installations with vertical wires of double that 

 length, i. e. 150 feet, were erected at a distance of 85 miles apart, and 

 signals were easih' o])tained between them. According to a rigorous 

 application of the law. 72 miles ought to have been obtained instead of 

 85; but as I have previously stated, the law has been proved only to be 

 approximately correct, the tendenc}' being alwavs on what I might call 

 the right side; thus we obtain a greater distance than the application 

 of the law would lead us to believe. There is a remarkable circum- 

 stance to be noted in the case of the 85 miles signaling. At the Alum 

 Bay station the mast is on the cliff, and there is no curvature of the 

 earth intervening between the two stations; that is to say, a straight 

 line ))etween the base of the Haven and Alum Ba}' stations would clear 

 the surface of the sea. But in the case of the 85 miles the two stations 

 were located on the sea level, and between them exists a hill of water, 

 owing to the earth's curvature, amounting to over 1,000 feet. If those 

 waves traveled only in straight lines, or the effect was noticeable onl}^ 

 across open space, in a direct line, the signals would not have been 

 received except with a vertical wire 1,000 feet high at both stations. 



While carrA^ing out some experiments nearh' three years ago at Salis- 

 bury, Captain Kennedy. R. E.. and I tried numerous forms of induction 

 coils wound in the ordinary way. that is, with a great number of turns 

 of wire on the secondary circuit, with the object of increasing, if pos- 

 sible, the distance or range of transmission; but in every case we 

 observed a very marked decrease in the distance obtainable with the 

 given amount of energy and height. Similar results were obtained 

 some months later, 1 am informed, in experiments carried out by the 

 general post-office engineers at Dover. 



In all our above-mentioned experiments the coils used were those in 

 which the primary consisted of a smaller or larger number of turns 

 of comparatively thick wire, and the ^econdar}^ of several la3"ers of 

 thinner wire. I believe I am right saying that hundreds of these 

 coils were tried, the result always be 4- that l)v their employment the 

 possible distance of signaling was considerably diminished instead of 

 being increased. We eventuall}^ found an entirely new form of induc- 

 tion coil that would work satisfactorily, and that began to increase the 

 distance of signaling. 



The results given by some of the new form of induction coils have 

 SM 1901 19 



